Skip to content
Sylvia van Os edited this page Jun 1, 2022 · 11 revisions

To read lines rather than words, pipe/redirect to a while read loop.

Problematic code:

for line in $(cat file | grep -v '^ *#')
do
  echo "Line: $line"
done

Correct code:

grep -v '^ *#' < file | while IFS= read -r line
do
  echo "Line: $line"
done

or without a subshell (bash, zsh, ksh):

while IFS= read -r line
do
  echo "Line: $line"
done < <(grep -v '^ *#' < file)

or without a subshell, with a pipe (more portable, but write a file on the filesystem):

mkfifo mypipe
grep -v '^ *#' < file > mypipe &
while IFS= read -r line
do
  echo "Line: $line"
done < mypipe
rm mypipe

NOTE: grep -v '^ *#' is a placeholder example and not needed. To just loop through a file:

while IFS= read -r line
do
  echo "Line: $line"
done < file
# or: done <<< "$variable"

Rationale:

For loops by default (subject to $IFS) read word by word. Additionally, glob expansion will occur.

Given this text file:

foo *
bar

The for loop will print:

Line: foo
Line: aardwark.jpg
Line: bullfrog.jpg
...

The while loop will print:

Line: foo *
Line: bar

Exceptions

If you do want to read word by word, you can set $IFS appropriately and disable globbing with set -f, and then ignore this warning. Alternatively, you can pipe through tr ' ' '\n' to turn words into lines, and then use while read. In Bash/Ksh, you can also use a while read -a loop to get an array of words per line.

ShellCheck

Each individual ShellCheck warning has its own wiki page like SC1000. Use GitHub Wiki's "Pages" feature above to find a specific one, or see Checks.

Clone this wiki locally